Download 2b_52_tropical lowland rainforest birds.PMD

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Conservation movement wikipedia , lookup

Bifrenaria wikipedia , lookup

Habitat conservation wikipedia , lookup

Forest wikipedia , lookup

Tropical rainforest wikipedia , lookup

Old-growth forest wikipedia , lookup

Operation Wallacea wikipedia , lookup

Reforestation wikipedia , lookup

Biological Dynamics of Forest Fragments Project wikipedia , lookup

Transcript
Tropical lowland rainforest birds
on a highly urbanized island:
monitoring, losses and lessons
Navjot S. Sodhi, Marjorie Castelleta, Wen Zsa Loh, Benjamin P. Y-H. Lee and R. Subaraj
Abstract
We studied tropical lowland rainforest birds in Singapore. Forest loss in Singapore, primarily due
to urbanization, has been heavy, with 95% of the native forest being cleared. We found that between 1923 and 1998, 61 (67%) out of 91 forest dependent bird species (species that occur only
within primary/secondary rainforest) have gone extinct. Based on observations of 2 forest dependent bird species (the Asian Fairy-Bluebird Irena puella and Short-tailed Babbler Malacocincla
malaccensis), we found that these species occupied remnants that were larger and less isolated
than unoccupied remnants. Artificial nest experiments revealed that predation on artificial nests
(80%) was higher in Singapore than reported by similar studies done elsewhere. Detailed studies in
2 larger remnants showed that the abundance of some bird species seemed to be affected by
factors such as food abundance and vegetation structure. However, bird diversity in these 2
remnants was poor and different from other larger forests within Southeast Asia. We recommend
that existing forest remnants within Singapore should be preserved so that the remaining level of
bird biodiversity is maintained. Large forest reserves within Southeast Asia should be preserved
so that forest birds in other countries do not face the same fate as in Singapore.
INTRODUCTION
Human activities such as urbanization and logging
have already cleared about 24% of all tropical
lowland rainforests (FAO 1993). Because of this
heavy deforestation, many tropical lowland
rainforests now occur in small patches (remnants or
fragments) (Turner and Corlett 1996). This heavy
rainforest loss and fragmentation can be detrimental
to biodiversity because of factors such as negative
edge effects (e.g. high nest predation in fragmented
forest) and reduced habitat quality and quantity.
Considering that a large number of endemic bird
species could go extinct because of heavy deforestation (Brooks et al. 1997), it is important to monitor
birds in tropical lowland rainforest remnants so that
the existing biodiversity can be maintained.
Singapore (103o50’E, 1o20’N) is an island state
containing the main island of 584 km2 and 60 small
offshore islands (total area 52 km2). Singapore
experiences typical equatorial climate with mean
annual rainfall of 2375 mm. Singapore has one of the
highest human population densities on our planet
with 6615 people/km2. Due to heavy urbanization,
Singapore offers one of the worst-case scenarios for
forest destruction with 95% of native forest being
cleared with the existing forest remnants scattered
throughout the island (Turner et al. 1994). Therefore,
Singapore provides an ideal location to determine
the effect of tropical rainforest loss due to heavy
urbanization on native forest bird species of Southeast Asia. Our overall objectives were to determine:
1) the extent of loss of forest bird species from
Singapore, 2) current bird diversity level in
rainforest remnants of Singapore and compare this
diversity with similar forests within Southeast Asia,
3) the possible factors (forest area, forest isolation,
forest structure and food resources) affecting forest
bird diversity, abundance and occurrence within
Singapore, and 4) demographics of forest bird
species within Singapore. We hope that with this
study, we will be able to make precise recommendations for the conservation of forest bird species of
Singapore. Here we present preliminary results from
our research.
Authors’ address: Department of Biological Sciences, National University of Singapore, Kent Ridge, Singapore
119260, Republic of Singapore
College of Agriculture and Life Sciences
78
METHODS
To determine the extent of bird species loss from
Singapore, 20 fragments (9 - 795 ha) were surveyed.
Using transect surveys, a one and half year round
study (June 1997 to November 1998) was conducted
sampling every 2 months each of the 5 late secondary/primary forests (> 50 years old), 8 young
secondary forests (< 50 years old) and 7 woodland
forests (abandoned plantations). These data were
supplemented with mist-netting surveys. Based on
transect counts and mist-netting, a species list was
made for the forest birds of Singapore. This species
list was compared with that made by Chasen (1923).
Reliable species records present in Chasen’s list but
absent from our list were considered to be of species
gone locally extinct from Singapore. We used 2 other
lists to confirm extinctions (Gibson-Hill 1949, Lim
and Gardner 1997).
Habitat requirements of forest dependent bird species
Thirteen forest remnants (4 - 795 ha) were surveyed from 29 July to 5 October 1996 (toward the
end of peak breeding season in Singapore; Hails and
Jarvis 1987) to determine how remnant size and
isolation affect the distribution of forest dependent
and non-forest dependent species. A remnant is
defined as a patch of native vegetation around
which most or all of the original vegetation has been
cleared. In this study, forest remnants, patches, sites,
and fragments are used interchangeably. All 13
remnants contained secondary forest with different
dominant plant species; 2 of them also contained
primary forest. The study fragments were surrounded by a variety of edges such as track, road,
reservoir and pipeline.
Remnants were surveyed 3 times each to determine the presence or absence of each of the 4 target
bird species. Two species (the Asian Fairy-Bluebird
Irena puella and Short-tailed Babbler Malacocincla
malaccensis) were forest dependent species (only
occur within the primary/secondary rainforest),
while 2 species (the Olive-winged Bulbul Pycnonotus
plumosus and Striped Tit-babbler Macronous gularis)
were forest non-dependent species (do not exclusively occur within the primary/secondary
rainforest). A single line transect (with unlimited
distance) was placed in each remnant along forest
trails, paths, or small roads in such a way to cover as
much area as possible. Target species were searched
for while walking slowly along the transect between
7:00 and 10:00 am on fair weather days (a period of 2
or more mornings was required to conduct surveys
in large fragments). All 13 remnants were surveyed
once without using playback of pre-recorded songs.
The subsequent 2 surveys involved song playback to
confirm absences. The songs of the 4 target species
were played in a random order at every 100-m
interval along the transect. The song of each species
was played for 2 min followed by a waiting period
of 1 min for that specific species to respond. Hence, a
total of 12 min was spent at every 100-m interval. A
target species was considered present if visual or
auditory contact was made.
Patch area and isolation distance were determined
for each remnant. These 2 variables describe the
remnants in relation to the surrounding landscape.
To determine the isolation of a remnant, distance to
the nearest forest remnant was measured from all 4
cardinal directions (north, south, east and west) and
a mean was calculated from these 4 values (isolation
distances for the study remnants ranged from 75 1510 m).
Artificial nest predation experiment
We used 5 forest fragments (2 primary forests, 2
secondary forests, and 1 abandoned rubber (Hevea
brasiliensis) plantation) within Singapore. For the
nest predation experiment, chicken eggs were used.
The use of chicken eggs served to simulate the eggs
of the Red Jungle Fowl (Gallus gallus), a groundnesting bird which occurs in Singapore. Fresh
chicken eggs (bought on the same day of experiment
set-up) were used. In addition to real eggs, plasticine
eggs were used to determine the type of predators
that attacked the eggs by the bite-marks left on the
plasticine. The imitation eggs were hand-made from
brown plasticine. The plasticine chicken eggs were
then painted with poster colors to give a ‘chickenegg’ color. During the handling of the real and
plasticine eggs, rubber gloves were worn to reduce
contamination with human scent.
The number of transects at the sites ranged from
15 to 20. Each transect started from the edge of the
forest (here defined as the first tree or shrub) and ran
100 m into the forest. The egg experimental stations
(containing 1 real chicken egg and 1 plasticine egg)
were located at the 0 m, 50 m and 100 m positions.
The experimental stations were placed at random
distances (1-5 m) and at random directions (left or
right) from the transect. To facilitate relocation, they
were usually placed near buttresses of trees or the
base of larger saplings. We attempted to cover the
eggs by adjusting the vegetation, such that when
viewed directly from above, at a height of 1.8 m, the
percentage cover directly above the eggs was about
50%. Although no attempt was made to construct an
artificial nest, we designated the egg experiment
stations as artificial nests because many groundnesting bird species do not build elaborate nests. The
transects within a fragment were spaced 50 m from
each other. Paths were not cleared along the
transects and every effort was made to minimize the
impact of setting up the experiment. The experi-
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004
79
ments were performed in August 1996, which is
during the end of the peak breeding season in
Singapore. To avoid pseudoreplication (Hurlbert
1984), we conducted experiments only once at each
site.
To determine the predation rates, experimental
nest stations were checked on day 8. Nests were
considered as preyed upon if: (i) an egg was missing,
(ii) there were cracks or peck marks on the real
chicken egg, (iii) the chicken egg was smashed, or
(iv) bite-marks were on the plasticine eggs. Predation by small mammals and other predators was
determined by examining the bite marks imprinted
on the plasticine eggs.
Bird biodiversity in 2 remnants
Detailed studies were conducted in 2 remnants; the
MacRitchie forest (521 ha) included primary forest
interspersed with secondary forest and the Nee Soon
forest (795 ha) included mostly secondary forest
embracing 15 ha of disturbed freshwater swamp
forest.
Bird census techniques — Line transects of approximately 4.5 km were selected for each site. The
transects were selected such that they spanned the
length of the remnant. Each transect was surveyed 8
times from 7:00 to 9:30 am, between 14 July and 24
September 1997. The surveys were carried out in the
morning as maximum diurnal bird detectability is
observed around dawn (Shields 1979). Two mornings
were needed to cover each transect in order to obtain
an optimal walking speed of about 1 km/h. All birds
seen or heard were recorded. Surveys were only
conducted on days with fine weather when there
would be maximal bird activity.
In addition, mist-netting was carried out 3 mornings
during each of the following 5 months: March, May,
July, August and September 1997. The advantage of
mist-netting is that it may result in trapping bird
species that are inconspicuous and/or seldom vocalize. The mist nets (12 x 2.6 m, 30 mm mesh) were set
up by about 7:00 am and checked every 45 min, until
they were taken down at 12:30 pm. All birds trapped
were identified using field guides. Mist net hours
conducted in MacRitchie and Nee Soon were 760.5
and 960.5, respectively. The birds were grouped
according to their diets (carnivores, insectivores,
frugivores or nectarivores) (Kang and Hails 1995). All
birds were also classified by vertical feeding zones
following the system used by Gregory-Smith (1996).
Arthropod sampling — Foliage along the transects
was sweep-sampled for the presence of understory
arthropods. Each transect was sampled for arthropods
on 5 different occasions. Sweep-sampling was conducted at intervals of 150 m to obtain a minimum
sample size of 30 sweep-samples per day of sampling.
Each sweep-sample consisted of 16 sweeps using a
80
butterfly net with a mesh-size small enough to prevent
the escape of most arthropods. Sweep-sampling in
each subsequent sampling occasion began 20 m from
the start of the previous sample in order to evenly
sample each transect.
Vegetation sampling— Vegetation was sampled at
intervals of 150 m along each transect (MacRitchie 35
plots and Nee Soon 32 plots). Each 5 m radius plot
was randomly located either to the left or right of the
transect by tossing a coin. In each plot, the diameter at
breast height (dbh) of all trees was measured using
dbh tape. A tree is defined as any plant at least 2 m in
height possessing a woody trunk. The trees in the plots
were counted and classified as understory trees (< 10
m tall) or medium story/canopy trees (> 10 m tall)
(Wong 1986). The percentage of canopy cover was
measured using a spherical densiometer (Lemmon
1957). Percent shrub cover was visually estimated by 2
independent observers. By inserting a ruler vertically
into the leaf litter at 13 fixed points in each plot, leaf
litter depth was measured and an average was
calculated. The number of fruiting trees along each
transect was surveyed 2 times (late August and late
September) and grouped as understory or canopy
trees.
Comparison with other forested areas in the region —
The bird diversity of MacRitchie and Nee Soon forests
was compared with other lowland forests in the region
to determine avian species richness. Nine forested
areas, including continuous forests and remnants,
were chosen from Bransbury (1993). To determine
similarities in species composition among the forests, a
cluster analysis using the program PC-ORD (McCune
and Mefford 1995) was carried out on data matrix
consisting of all 11 sites (including MacRitchie and
Nee Soon) x 246 forest bird species.
Data analyses — Data analyses were carried out using
various statistical programs (e.g. StatView SE +
Graphicsä) following standard statistical techniques.
Statistical significance for all tests was set at a = 0.05.
In text and table, + refers to standard error.
RESULTS
Local extinctions and habitat requirements of
selected bird species
Between 1923 and 1998, of 91 forest dependent
bird species, 61 (67%) have gone extinct from
Singapore. Patch area and isolation affected the
distribution of forest dependent species. Non-forest
dependent species, the Striped Tit-babbler and
Olive-winged Bulbul occurred in 13 (100% of the
remnants surveyed) and 9 remnants, respectively.
The Asian Fairy-Bluebird and Short-tailed Babbler,
both forest dependent species, occupied the same 4
remnants. The remnants occupied by these 2 species
were larger (537.75+251.68 ha) than unoccupied
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004
remnants (32.22+4.89 ha; Mann-Whitney U = 36, df =
9, 4, P = 0.006). Similarly, the remnants occupied by
these 2 species were less isolated (341.25+109.06 m)
than unoccupied remnants (960.00+151.79; U = 31.5,
df = 9, 4, P = 0.04). However, the areas and isolation
of occupied remnants did not differ significantly
from those of unoccupied remnants for the Olivewinged Bulbul (Mann-Whitney U-tests, P > 0.09).
Artificial nest predation rates
The predation rate on artificial ground nests for all
the 5 fragments combined was 80% (264/328 nests).
Much of this predation was by small mammals
(55%; 146/264 nests). As mentioned, of the 5 fragments, 2 contained primary forests (artificial nest
predation rates: 68% and 91%), 2 contained secondary forests (80% and 91%) and 1 contained abandoned rubber plantation (86%). Predation rates
differed between the primary forests (G = 6.94, df =
1, P = 0.01). Predation rates did not differ between
secondary forests and among these 2 forests and
rubber plantation (G-tests, P > 0.11). Secondary
forests and the rubber plantation combined had
higher predation rates than 1 of the primary forests
(G = 6.0, df = 1, P = 0.01) but not the other (G = 1.06,
df = 1, P = 0.30). No significant differences were
detected in nest predation rates relative to the
distance from the edge in any of the fragments
(Kruskal-Wallis ANOVAs, P > 0.16).
Bird communities of 2 remnants
Plots of the cumulative number of species recorded with each successive survey revealed that
adequate sampling was done both for transect
counts and mist-netting in both MacRitchie and Nee
Soon. Based on transect counts, the mean number of
species recorded was 29.0+1.97 and 25.5+1.24 in
MacRitchie and Nee Soon, respectively (U = 21.5, df
= 8,8, P = 0.26). The mean number of individuals
counted was 380.1+29.26 and 362.5+35.3 in
MacRitchie and Nee Soon, respectively (U = 26, df =
8,8, P = 0.52). Forty-three and 40 bird species were
recorded in MacRitchie and Nee Soon, respectively
(Table 1). Thirty-nine species were common to the
both remnants; however, there were some differences in species occurrences and abundance between
the remnants (Table 1).
Mist-netting yielded 1 extra bird species for each
site; Emerald Dove (Chalcophaps indica) for
MacRitchie and White-chested Babbler (Trichastoma
rostratum) for Nee Soon. Forty and 51 bird species
were trapped in MacRitchie and Nee Soon, respectively. The retrapping rate was 12.5% and 11.8% for
MacRitchie and Nee Soon, respectively.
Data from the transect counts and mist-netting
were pooled for the following analyses. The mean
number of understory insectivore bird individuals
(White-rumped Shama Copsychus malabaricus, Darknecked Tailorbird Orthotomus atrogularis, Abbott’s
Babbler M. abbotti, Short-tailed Babbler and Chestnut-winged Babbler Stachyris erythroptera) did not
differ significantly between MacRitchie (73.63+63)
and Nee Soon (87.88+7.57; U = 19, df = 8,8, P = 0.17).
The mean number of arthropods also did not differ
between the 2 forests (11.8+0.39 and 9.66+0.67; U = 4,
df = 5,5, P = 0.07). More frugivore bird individuals
(Red-crowned Barbet Megalaima rafflesii, Long-tailed
Parakeet Psittacula longicauda and Pink-necked
Green-Pigeon Treron vernans) were recorded in
MacRitchie (maximum number = 45) than in Nee
Soon (10; c2 = 22.27, df = 1, P = 0.0001). More fruiting
trees were available in MacRitchie (maximum
number = 352) than in Nee Soon (285; c2 = 7.06, df =
1, P = 0.007).
There were more canopy trees in MacRitchie (560)
than in Nee Soon (326; c2 = 61.8, df = 1, P = 0.0001).
Based on transect counts and mist-netting, more
individuals of bird species strictly foraging in the
canopy and upper story (e.g. Hill Myna Gracula
religiosa) were found in MacRitchie (286) than in Nee
Soon (165) (c2 = 32.46, df = 1, P = 0.0001). There was
higher shrub cover in Nee Soon (70.0+4.47) than in
MacRitchie (54.36+4.61; U = 363.5, df = 35,32, P =
0.01). Based on transect counts and mist-netting,
more individuals of understory species (e.g. Chestnut-winged Babbler) were found in Nee Soon (163)
than in MacRitchie (102; c2 = 14.04, df = 1, P =
0.0002). None of the other vegetation variables
recorded seemed to affect the bird fauna.
The cluster analysis showed that the bird communities of MacRichie and Nee Soon were most similar
to each other and similar to those found on the
island of Pulau Tioman. Many of the larger continuous forests such as Taman Negara (n = 182 bird
species) and Endau-Rompin (n = 175) had the most
different bird communities from MacRitchie and
Nee Soon.
DISCUSSION
A large proportion of forest dependent birds have
gone extinct from Singapore (e.g. all members of
families Trogonidae, Bucerotidae and Eurylamidae).
There may be many factors causing the extinction of
forest dependent bird species in Singapore. The most
important may be the habitat loss, followed by
illegal trapping and competition from exotic bird
species (e.g. Javan Myna Acridotheres tristis). Some of
the current forest depend bird species may exist in
small numbers and may go extinct in the future. For
example, there is only 1 pair of the White-bellied
Woodpecker (Dryocopus javensis) currently surviving
in Singapore. Therefore, extinctions of forest dependent bird species probably will continue unless
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004
81
concrete conservation measures are implemented.
Habitat requirements of Southeast Asian forest
dependent bird species have been poorly documented. We found that the remnants in which 2
forest bird species (the Asian Fairy-Bluebird and
Short-tailed Babbler) occurred were larger and less
isolated than unoccupied remnants. The territory
sizes and feeding ranges of these species are not
known, but may be related to their need for relatively large remnants. The 4 remnants that these 2
species occupied had been connected until the early
1990s. This may have affected the occupancy of these
particular remnants as they may have occurred
across the contiguous area prior to isolation.
Results of artificial nest predation experiments
showed that predation rates were relatively high in
Singapore compared to similar studies done elsewhere (Yahner and Wright 1985, Gibbs 1991, Burkey
1993). The unusually high artificial nest predation
rates in Singapore could be due to several reasons.
First, matrix (area surrounding remnants) is highly
urbanized and may harbor relatively high densities
of potential nest predators such as feral animals (e.g.
cats) and corvids. Second, due to heavy loss of
native forests, many potential predators (e.g. monkeys) may be confined only to remaining pockets of
forests. Last, it is possible that the use of artificial
painted eggs may have resulted in high levels of
predation of experimental nests.
The bird communities of MacRitchie and Nee Soon
were broadly similar despite some differences in
forest composition (e.g. primary forest in
MacRitchie). Some food resources (fruiting tree
availability) seemed to affect the abundance of
frugivores. Similar results have been previously
reported (Wong 1986). Unlike some previous studies
(Janzen 1973, Wong 1986), we failed to find a significant relationship between insectivores and arthropod abundance. This may be due to our small
sample size. Some variables of vegetation structure
(e.g. shrub cover) affected the abundance of some
species. We found that the bird diversity in
Singapore’s 2 larger remnants was poor compared to
other similar but larger forests in Southeast Asia.
Larger forests may contain diverse species composition because of factors such as high quantity and
quality (high habitat heterogeneity and low human
disturbance) of existing forests.
This study shows that many forest dependent bird
species have gone extinct from Singapore and that
existing avifauna is depauperate compared to other
larger forests within Southeast Asia. We recommend
that existing native forests in Singapore be preserved
so that the current level of biodiversity is maintained. Large forest reserves within Southeast Asia
should be maintained so that forest birds in other
countries do not face the same fate as in Singapore.
82
ACKNOWLEDGMENTS
This study was supported by the National University of Singapore (RP960316) and the Cheng-Kim
Loke Foundation. S. L. Huong, D. Yap, M. Soh, K.
Ong, B. Fitz-Earle, T. Tan, L. Ong and L. H. Liow
provided assistance. The National Parks Board and
Primary Production Department provided the
permits. We thank them all.
REFERENCES
Bransbury, J. 1993. A birdwatcher’s guide to Malaysia. Waymark Publishing, Singapore. 282 pp.
Brooks, T. M., S. L. Pimm and N. J. Collar. 1997.
Deforestation predicts the number of threatened
birds in insular Southeast Asia. Conservation
Biology 11:382-394.
Burkey, T. V. 1993. Edge effects in seed and egg
predation at two neotropical rainforest sites.
Biological Conservation 66:139-143.
Chasen, F. N. 1923. An introduction to the birds of
Singapore Island. Singapore Naturalist 2:87-112.
FAO. 1993. Forest resources assessment 1990:
tropical countries. FAO forestry paper 112. 61 pp.
Gibson-Hill, C. A. 1949. A checklist of the birds of
Singapore island. Bulletin of Raffles Museum
21:132-183.
Gibbs, J. P. 1991. Avian nest predation in tropical wet
forest: an experimental study. Oikos 60:155-161.
Gregory-Smith, R. 1996. A pocket checklist of the
birds of Peninsular Malaysia and Singapore.
Malaysian Nature Society, Kuala Lumpur, Malaysia. 44 pp.
Hails, C. J., and F. Jarvis. 1987. Birds of Singapore.
Times Edition, Singapore. 168 pp.
Hurlbert, S. H. 1984. Pseudoreplication and the
design of ecological field experiments. Ecological
Monograph 54:187-211.
Janzen, D. H. 1973. Sweep samples of tropical foliage
insects: effects of seasons, vegetation types,
elevation, time of day and insularity. Ecology
54:687-708.
Kang, N., and C. J. Hails. 1995. The birds of the Bukit
Timah Nature Reserve and Central Catchment
Area. Garden’s Bulletin Supplement 3:121-143.
Lemmon, P. E. 1957. A new instrument for measuring forest overstorey density. Journal of Forestry
55:667-668.
Lim, K. S., and D. Gardner. 1997. Birds: an illustrated
field guide to the birds of Singapore. Sun Tree
Publishing Limited, Singapore. 226 pp.
McCune, B., and M. J. Mefford. 1995. PC-ORD.
Multivariate analysis of ecological data, version
2.0. MjM Software Design, Gleneden beach,
Oregon. 126 pp.
Shields, W. 1979. Avian census techniques: an
analytic review. Pages 23-51 in J. G. Dickson, R. N.
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004
Conner, R. R. Fleet, J. C. Kroll and J. A. Jackson,
eds. The role of insectivorous birds in forest
ecosystems. Academic Press, New York.
Turner, I. M., and R. Corlett. 1996. The conservation
value of small, isolated fragments of lowland
tropical rain forest. Trends in Ecology and Evolution 11:330-333.
Turner, I. M., H. T. W. Tan, Y. C. Wee, A. B. Ibrahim,
P. T. Chew and R. T. Corlett. 1994. A study of plant
species extinction in Singapore: lessons for the
conservation of tropical biodiversity. Conservation
Biology 8:705-712.
Wong, M. 1986. Tropical organization of understory
birds in a Malaysian dipterocarp forest. Auk
103:100-116.
Yahner, R. H., and A. L. Wright. 1985. Depredation
on artificial ground nests: effects of edge and age
plot. Journal of Wildlife Management 49:508-513.
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004
83
84
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004
85
86
Proceedings 4th International Urban Wildlife Symposium. Shaw et al., Eds. 2004